Can the following question be posted? “Where could an error be in the proposed proof of Fermat Last Theorem? Info about the author, proof and preliminary validation is at (link)”. How should the question be reformulated to not be deleted? Previously, I tried to post a similar question with more details, but it was deleted.
That is off-topic, as is asking for correctness of claimed proofs about any famous conjecture, especially if, as I guess, it's a simple proof and the link is at viXra or ResearchGate or Google Drive/Dropbox etc. Even if it's on the arXiv, that's not going to get a free pass, since people get claimed but completely broken proofs of famous conjectures not infrequently. You should think of MathOverflow as something like a mathematics department common room. If you wouldn't barge in and start pushing a copy of your paper on everyone present asking them to check it now, then it's not suitable here. And if someone did come barging in to the common room in this way, they would be shown the door as fast as possible, which is what happens here.
Even if someone comes asking about the correctness of a random preprint they found (on Twitter or Reddit of Facebook or whatever), it's off-topic. Sometimes people come asking for such verification of their own work, but pass it off as something they just happened across, asking here under a pseudonym. Such behaviour is clearly not going to help.
There is a long history of such a policy for this and similar situations, for example:
And in case anyone thinks that this is a gate-keeping issue of the mathematics community keeping people out, see the discussion here:
where the matter of a preprint with a claimed proof of a famous open problem by a famous mathematician needed to be dealt with. The tl;dr is that it is still off-topic.